Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Feb 4.
Published in final edited form as: Org Lett. 2011 Jan 6;13(3):450–453. doi: 10.1021/ol102798f

Enantioselective Synthesis of 2,2,5-Tri- and 2,2,5,5-Tetrasubstituted Tetrahydrofurans via [4+2] Cycloaddition and Ring-Opening Cross-Metathesis

Noah M Benjamin 1, Stephen F Martin 1,*
PMCID: PMC3076641  NIHMSID: NIHMS263035  PMID: 21210696

Abstract

graphic file with name nihms-263035-f0001.jpg

A chiral vinyl sulfoxide has been developed that undergoes highly diastereoselective Diels-Alder cycloadditions with various substituted furans in excellent yield. The cycloadducts can be stereoselectively transformed into 2,2,5-tri- and 2,2,5,5-tetrasubstituted tetrahydrofurans, which are structural subunits of many natural products, via regioselective ring-opening metathesis/cross-metathesis or oxidative cleavage/refunctionalization.


The 2,2,5-tri- and 2,2,5,5-tetrasubstituted tetrahydrofuran substructure 1 is a common motif found in a number of biologically active natural products such as cortistatin A,1 (+)-davanone,2 and caruifolin A3 (Figure 1). In the context of several ongoing projects, we were confronted with the challenge of the enantioselective synthesis of substructures related to 1 (Z = H). A survey of the literature4 revealed that available methodology for such constructions is limited largely to cyclizations of unsaturated or epoxy alcohols, and more general entries to these important substructures are lacking.

Figure 1.

Figure 1

Natural products related to 1.

Various enantioselective [4+3]5 and [4+2]6,7 cycloadditions involving furan itself to give bridged bicyclic adducts are known, but the application of such tactics for elaboration to 2- and 2,5-substituted furans is not well established. Moreover, tactics for the elaborating the products of these cycloadditions into monocyclic, substituted tetrahydrofurans are not well developed.8,9 In view of the current state of the art, we recognized the significant opportunity to develop an efficacious entry to highly substituted tetrahydrofurans related to 1, and we now report the results of some of our findings.

Based upon a survey of the literature, we reasoned that processing enantiomerically pure oxabicycloheptenes 3 via regioselective ring-opening cross-metathesis (ROCM) or oxidative cleavage would lead to tetrahydrofurans 2;8,9 removal or refunctionalization of the EWG in 2 would then deliver the desired tetrahydrofurans 1 (Scheme 1). Access to enantiomerically pure 3 would require the development of a highly enantioselective or diastereoselective Diels-Alder reaction of substituted furans 5 with a dienophile 4 wherein the electron withdrawing group (EWG) could be easily transformed into a hydrogen atom or an oxygen function. Namely, we required that 4 serve as an ethylene, vinyl alcohol or ketene equivalent. We were thus attracted to the use of sulfoxides as the preferred activating groups because this moiety can be more readily removed and/or refunctionalized than carboxylic acid derivatives.10 p-Tolyl vinyl sulfoxide (6a) was known to undergo [4+2] cycloaddition with furan, but not substituted furans, in the presence of a Lewis acid promoter in good yield and de,7 so we initiated our studies by examining the reaction of 6a with 2,5-dimethylfuran (7c). After screening a series of Lewis-acids, we found TBSOTf to be the optimal promoter for this cycloaddition giving oxabicycle 8a in 68% yield (Table 1). In an effort to improve the yield in this reaction, we explored other aryl vinyl sulfoxides as dienophiles and discovered that cycloadditions of 7c with phenyl vinyl sulfoxide (6b) and p-chlorophenyl vinyl sulfoxide (6c) proceeded to give 8b and 8c in 81% and 94% yields, respectively.

Scheme 1.

Scheme 1

Retrosynthetic analysis of 1

Table 1.

Development of Dienophile 6c

graphic file with name nihms-263035-t0007.jpg
entry Ar isolated yield (%)a
a Tolyl 68
b Phenyl 81
c 4-Chlorophenyl 94
a

all cycloadditions proceeded with greater than >20:1 endo:exo selectivity.

Having established that 6c was the preferred dienophile, it was necessary to develop an efficient means to prepare both enantiomers in pure form. Fortunately, we were able to adapt a procedure that had been developed by Maignan for the preparation of both enantiomers of 6a11 (Scheme 2). In the event, racemic 6c underwent base-induced, conjugate addition of (−)-menthol (8) giving the easily separable diastereomers 10 and 11; the structure of 10 was determined by X-ray crystallographic analysis. Elimination of mentholate ion was then promoted using LiHMDS, giving both (−)-6c and (+)-6c in >99% ee and good yield. Unlike a previous report of a similar reaction,11 our procedure does not rely upon methylation of the intermediate alkoxide, thereby allowing facile recovery of menthol in about 90% yield.

Scheme 2.

Scheme 2

Enantioselective Synthesis of (+)-6c and (−)-6c

We then examined the diastereoselectivities of the cycloadditions of enantiopure (+)-6c with furans 7a-c in the presence of TBSOTf, and the results of the optimized reactions are collected in Table 2. Two workup procedures were developed that enabled the direct and facile synthesis of either sulfinyl substituted cycloadducts 12a–c or sulfonyl substituted adducts 13a-c. The ability to access both cycloadducts with equal ease enhances the versatility of the method. It is noteworthy that the yield and diastereoselectivity observed for the reaction of (+)-6c with furan (7a) compare favorably with the report of Kagan for the reaction of 7a with (−)-tolyl vinyl sulfoxide. The stereoselectivity of the cycloaddition improves significantly when the substituted furans 7b,c are employed as dienes. The structures of 12a, 12b, and 13c were established by X-ray crystallography.

Table 2.

Diels-Alder Cycloaddition with Substituted Furans

graphic file with name nihms-263035-t0008.jpg
furan product L.A.
equiv
furan
equiv
temp
(°C)
yield
(%)
endo/exoa endo de/ee
(%)b
6a 12a 1 5 0 98 1.8/1 90
6a 13a 1 5 0 98 - 88
6b 12b 0.75 2 −30 89c 25/1 99
6b 13b 0.75 2 −30 81c - 99
6c 8c 0.50 2 −30 94 25/1 98
6c 13c 0.50 2 −30 88 - 98
a

determined by NMR.

b

determined by chiral HPLC.

c

cycloaddition proceeded with complete regioselectivity, with methyl group proximal to sulfoxide/sulfone.

Having developed the highly stereoselective cycloadditions of various furans with the chiral vinyl sulfoxide 5c, it was then necessary to identify ring-opening transformations that would provide 2,2,5-tri- and 2,2,5,5-tetrasubstituted tetrahydrofurans related to 1. Toward this end, we were attracted to the possibility that regioselective, ring-opening/cross-metathesis (ROCM) of the oxabicyclic systems 13a-c might lead to 2,2,5-tri- and 2,2,5,5-tetrasubstituted tetrahydrofurans of the general form 1. This premise was founded on the knowledge that regioselective ROCM reactions had been applied to unsymmetrically substituted norbornene derivatives bearing electron-withdrawing groups.9,12 Consistent with our expectations, reaction of enantiomerically pure 13c with excess allyltrimethylsilane in the presence of Hoveyda-Grubbs 2nd generation catalyst13 led to a completely regioselective ROCM reaction to furnish 14a as the only observable product in 94% yield (Table 3). In order to expand the scope of this ROCM, a number of monosubstituted alkenes were allowed to react with racemic 13c to give racemic 14b–f as the only observed products in yields ranging from 55-93%. The structure of 14a was determined by X-ray analysis, whereas those of 14b-f were assigned by a characteristic nOe signal between the vinylic protons and the proximal methyl substituents. Although several of these ROCM reactions proceeded with identical regioselectivity using the sulfoxide 12c, >10% catalyst loadings were required, and the yields were generally ≈30% lower.

Table 3.

Ring-Opening/Cross-Metathesis of 13c

graphic file with name nihms-263035-t0009.jpg
entry graphic file with name nihms-263035-t0010.jpga,b cat. loading
(mol %)
temp
(°C)
product yield
(%)
a graphic file with name nihms-263035-t0011.jpg 2.5 25 14a 94
b graphic file with name nihms-263035-t0012.jpg 5 25 14b 93
c graphic file with name nihms-263035-t0013.jpg 5 0 14c 89
d graphic file with name nihms-263035-t0014.jpg 10 25 14d 55c
e graphic file with name nihms-263035-t0015.jpg 2.5 25 14e 82
f graphic file with name nihms-263035-t0016.jpg 5 25 14f 78
a

6-10 equiv olefin.

b

Reaction performed with enantiomerically pure 13c for entry a and racemic 13c for entries b-f.

c

Reaction only proceeds under an atmosphere of ethylene.

We then queried whether regioselective ROCM reactions could be applied to 13b. Reaction of enantiomerically pure 13b with allyltrimethylsilane in the presence of Hoveyda-Grubbs 2nd generation catalyst at 70 °C proceeded regioselectively to give a mixture (12:1) of 15a and 16a in 76% yield (Table 4). Interestingly, when the reaction was performed at room temperature, the selectivity eroded, and a mixture (3:1) 15a and 16a was isolated. Several other olefins also engaged in efficient ROCM with 13b, but the reaction with methyl acrylate proceeded in the opposite regiochemical sense. The origin of this difference is not clear, but the reversal of regiochemical preference in ROCM reactions based on the electronics of the metathesis cross-partner has been observed previously.12 When 13a was used as a substrate in these ROCM reactions, only mixtures were obtained.

Table 4.

Ring-Opening/Cross-Metathesis of 13b

graphic file with name nihms-263035-t0017.jpg
entry graphic file with name nihms-263035-t0018.jpga,b cat. loading
(mol %)
temp
(°C)
yield
(%)
15:16
a graphic file with name nihms-263035-t0019.jpg 1 70 76 12:1
b graphic file with name nihms-263035-t0020.jpg 5 70 67c 0:1
c graphic file with name nihms-263035-t0021.jpg 1 84 59 1:0
d graphic file with name nihms-263035-t0022.jpg 2.5 25 99 -
a

6-10 equiv olefin.

b

Reaction performed with enantiomerically pure 13c for entry a and racemic 13c for entries b-d.

c

Reaction was run in neat methyl acylate in a sealed tube.

Oxidative cleavage was then explored as a tactic to convert oxabicyclo[2.2.1]heptenes into substituted tetrahydrofurans. For example, we found that ozonolysis of 13c in the presence of lutidine at −78 °C in acetone/H2O (9:1),14,15 followed by reduction of the crude product with excess sodium borohydride furnished the diol 17 in 84% yield (Scheme 3). Alternatively, reaction of the intermediate obtained from ozonolysis with the Ohira-Bestmann reagent gave the diyne 18 in 53% yield as a mixture (4:1) of epimeric sulfones.16

Scheme 3.

Scheme 3

Oxidative Cleavage Methods

In order to access substituted tetrahydrofurans of the general type 1, it remained to establish a suitable protocol for removing the sulfonyl group. Although there are methods for converting sulfones into ketones by oxidative desulfonylation to give tetrahydrofurans related to 1 (Z = O, OH),10d we were primarily interested in targeting substituted tetrahydrofurans of the general type 1 (Z = H). Accordingly, we found after extensive experimentation that the sulfonyl group in 14a could be selectively removed without concommitant carbon-oxygen bond cleavage by the action of Na-Hg amalgam in aqueous methanol to deliver 19 in 79% yield (Scheme 4).

Scheme 4.

Scheme 4

Desulfurization of 14a

In summary, we have developed the chiral vinyl sulfoxide 6c, both enantiomers of which are readily available, that undergoes highly diastereoselective [4+2] cycloadditions with several substituted furans. These represent the first examples of the use of substituted furans in enantioselective [4+2] cycloadditions with vinyl sulfoxides. The cycloadducts obtained from the reaction of 6c with furans may be further processed via several different refunctionalization manifolds to give facile access to 2,2,5-tri- and 2,2,5,5-tetrasubstituted tetrahydrofurans that comprise structural subunits in a broad array of biologically active natural products. We are also exploring other applications of 6c as a chiral dienophile in other Diels-Alder reactions.

Supplementary Material

1_si_001

Acknowledgment

We thank the National Institutes of Health (GM 31077) and the Robert A. Welch Foundation (F-652) for their generous support of this research. We are also grateful to Dr. Richard Pederson (Materia, Inc.) for catalyst support and to Anna J. Smith (The University of Texas) for helpful discussions.

Footnotes

Supporting Information Available Experimental procedures, spectral data, and copies of 1H and 13C NMR spectra for all new compounds. This material is available free of charge via the Internet at http://pubs.acs.org.

References

  • (1).Aoki S, Watanabe Y, Sanagawa M, Setiawan A, Kotoku N, Kobayashi M. J. Am. Chem. Soc. 2006;128:3148–3149. doi: 10.1021/ja057404h. [DOI] [PubMed] [Google Scholar]
  • (2).Sipma G, Van der Wal B. Recl. Trav. Chim. Pays-Bas. 1968;87:715–720. [Google Scholar]
  • (3).Ma C, Nakamura N, Hattori M, Zhu S, Komatsu K. J. Nat. Prod. 2000;63:1626–1629. doi: 10.1021/np000005+. [DOI] [PubMed] [Google Scholar]
  • (4).(a) Figadere B, Harmange J. Tetrahedron: Asymmetry. 1993;4:1711–1754. [Google Scholar]; (b) Hay MB, Wolfe JP. Tetrahedron. 2007;63:261–290. doi: 10.1016/j.tet.2006.08.105. [DOI] [PMC free article] [PubMed] [Google Scholar]; (c) Lee HM, Nieto-Oberhuber C, Shair MD. J. Am. Chem. Soc. 2008;130:16864–16866. doi: 10.1021/ja8071918. [DOI] [PubMed] [Google Scholar]; (d) Morimoto Y, Okita T, Kambara H. Angew. Chem. Int. Ed. 2009;48:2538–2541. doi: 10.1002/anie.200805857. [DOI] [PubMed] [Google Scholar]; (e) Ward AF, Wolfe JP. Org. Lett. 2010;12:1268–1271. doi: 10.1021/ol1001472. [DOI] [PMC free article] [PubMed] [Google Scholar]; (f) Marshall JA, Griot CA, Chobanian HR, Myers WH. Org. Lett. 2010;12:4328–4331. doi: 10.1021/ol101776e. [DOI] [PubMed] [Google Scholar]
  • (5).(a) Kende AS, Huang H. Tet. Lett. 1997;38:3353–3356. [Google Scholar]; (b) Xiong H, Huang J, Ghosh SK, Hsung RP. J. Am. Chem. Soc. 2003;125:12694–1295. doi: 10.1021/ja030416n. [DOI] [PubMed] [Google Scholar]; (c) Harmata M, Ghosh SK, Hong X, Wacharasindhu S, Kirchhoefer P. J. Am. Chem. Soc. 2003;125:2058–2059. doi: 10.1021/ja029058z. [DOI] [PubMed] [Google Scholar]; (d) Huang J, Hsung RP. J. Am. Chem. Soc. 2005;127:50–51. doi: 10.1021/ja044760b. [DOI] [PubMed] [Google Scholar]; (e) Harmata M, Brackley JA, Barnes CL. Tetrahedron Lett. 2006;47:8151–8155. [Google Scholar]; (f) Olson JP, Davies HML. Org. Lett. 2008;10:573–576. doi: 10.1021/ol702844g. [DOI] [PubMed] [Google Scholar]; (g) Xu J, Caro-Diaz EJE, Theodorakis EA. Org. Lett. 2010;12:3708–3711. doi: 10.1021/ol1015652. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (6).(a) Fraile JM, Garcia JI, Gracia D, Mayoral JA, Pires E. J. Org. Chem. 1996;61:9479–9482. [Google Scholar]; (b) Hayashi Y, Nakamura M, Nakao S, Inoue T, Shoji M. Angew. Chem. Int. Ed. 2002;41:4079–4082. doi: 10.1002/1521-3773(20021104)41:21<4079::AID-ANIE4079>3.0.CO;2-N. [DOI] [PubMed] [Google Scholar]; (c) Ryu DH, Kim KH, Sim JY, Corey EJ. Tetrahedron Lett. 2007;48:5735–5737. [Google Scholar]
  • (7).Ronan B, Kagan HB. Tetrahedron: Asymmetry. 1991;2:75–90. [Google Scholar]
  • (8).For examples, see: Ager David J., East Michael B. J. Chem. Res. Synopses. 1986:462–463. Klein LL, Shanklin MS. J. Org. Chem. 1988;53:5202–5209.
  • (9).(a) Arjona O, Csákÿ AG, Murcia MC, Plumet J. J. Org. Chem. 1999;64:9739–9741. [Google Scholar]; (b) Usher LC, Estrella-Jimenez M, Ghiviriga I, Wright DL. Angew. Chem. Int. Ed. 2002;41:4560–4562. doi: 10.1002/1521-3773(20021202)41:23<4560::AID-ANIE4560>3.0.CO;2-B. [DOI] [PubMed] [Google Scholar]; (c) Arjona O, Csákÿ AG, Plumet J. Eur. J. Org. Chem. 2003:611–622. [Google Scholar]; (d) Rainier JD, Liu Z. Org. Lett. 2005;7:131–133. doi: 10.1021/ol047808z. [DOI] [PubMed] [Google Scholar]; (e) Ikoma M, Oikawa M, Gill MB, Swanson GT, Sakai R, Shimamoto K, Sasaki M. Eur. J. Org. Chem. 2008:5215–5220. doi: 10.1002/ejoc.200800704. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • (10).(a) Magnus PD. Tetrahedron. 1977;33:2019–2045. [Google Scholar]; (b) Brown AC, Carpino LA. J. Org. Chem. 1985;50:1749–1750. [Google Scholar]; (c) Becker S, Fort Y, Caubere P. J. Org. Chem. 1990;55:6194–6198. [Google Scholar]; (d) Najera C, Yus M. Tetrahedron. 1999;55:10547–10658. [Google Scholar]; (e) Clayden J, Kenworthy MN, Helliwell M. Org. Lett. 2003;5:831–834. doi: 10.1021/ol0340585. [DOI] [PubMed] [Google Scholar]
  • (11).Alexandre C, Guillot C, Hayes P, Maignan C. Tetrahedron Lett. 1998;39:5769–5770. [Google Scholar]
  • (12).(a) Carreras J, Avenoza A, Busto JH, Peregrina JM. J. Org. Chem. 2009;74:1736–1739. doi: 10.1021/jo802399k. [DOI] [PubMed] [Google Scholar]; (b) Weeresakare GM, Liu Z, Rainier JD. Org. Lett. 2004;6:1625–1627. doi: 10.1021/ol049561b. [DOI] [PubMed] [Google Scholar]
  • (13).Kingsbury JS, Harrity JPA, Bonitatebus PJ, Hoveyda AH. J. Am. Chem. Soc. 1999;121:791–799. [Google Scholar]
  • (14).The presence of tertiary amines is known to improve the yield of certain ozonolysis reactions. See: Smith AB, Kingery-Wood J, Leenay TL, Nolen EG, Sunazuka T. J. Am. Chem. Soc. 1992;114:1438–1449.
  • (15).Ozonolysis in the presence of water leads directly to aldehydes. See: Schiaffo CE, Dussault PH. J. Org. Chem. 2008;73:4688–4690. doi: 10.1021/jo800323x.
  • (16).Roth GJ, Liepold B, Muller SG, Bestmann HJ. Synthesis. 2004:59–62. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1_si_001

RESOURCES